ES2938734T3 - Procedure for the manufacture of analogues of GLP-1 - Google Patents

Abstract

The present invention provides a process for the manufacture of GLP-1 analogues with high yield and purity by condensation of fragments in solid phase. In particular, it describes a convergent synthesis by condensation of a C-terminal pseudoproline fragment A with a fragment B attached to a solid support, followed by deprotection and cleavage of the support and final purification to produce the desired peptide. The invention further provides intermediate products useful in the manufacturing process. (Automatic translation with Google Translate, without legal value)

Description

DESCRIPTION

Procedure for the manufacture of analogues of GLP-1

field of invention

The present invention relates to the synthesis of peptides. In particular, it relates to processes for the manufacture of GLP-1 analogues. More particularly, the present invention relates to a process for the manufacture of liraglutide and semaglutide by condensation of peptide fragments.

Background of the invention

Glucagon-like peptide-1 (GLP-1) is a naturally occurring peptide hormone with one of the most potent insulinotropic activities. GLP-1 is part of a larger peptide produced and processed into GLP-1 in the pancreas and intestine. Glucagon-like peptide-1 (GLP-1) receptor agonists, or GLP-1 analogues, are a relatively new group of injectable drugs for the treatment of type 2 diabetes. One of their advantages over secretagogues older insulin medications, such as sulfonylureas or meglitinides, is that they have a lower risk of causing hypoglycemia.

Liraglutide and semaglutide are both highly similar peptides 30 amino acids in length, differing only in the amino acid at position 2 (aa ² , which is alanine, Ala, in liraglutide and α-aminoisobutyric acid, Aib, in semaglutide) and on the side chain attached to lysine at position 20 (Lys ²⁰ ), ie, the lipophilic albumin-binding moiety. Attached to the s-amino group of such a lysine, liraglutide has a Glu-spaced palmitic acid (indicated as Pal-Glu), whereas semaglutide carries an acylated glutamic-octadecanoic acid plus an additional spacer. Liraglutide and semaglutide are represented by using the “three-letter code” for amino acids/peptides, as follows:

1 5 10 15 20

His-Xi-Glu-Gly-Thr-Phe-Thr-Ser-Asp-Val-Ser-Ser-Tyr-Leu-Glu-Gly-Gln-Ala-Ala-Lys(X ² )-25 30 31

Glu-Phe-Ile-Ala-Trp-Leu-Val-Arg-Gly-Arg-Gly

wherein amino acid residue numbering begins with 1, indicating N-terminal histidine, and ends with 31, indicating C-terminal glycine,

and in which

in liraglutide (SEQ ID NO: 1)

X ¹ is Wing and

X2 is N-(1-oxohexadecyl)-L-y-glutamyl (also denoted as Pal-Glu), pictured here below

and

in semaglutide (SEQ ID NO: 2)

X1 is Aib (α-aminoisobutyric acid) and

X2 is N-(17-carboxy-1-oxoheptadecyl)-Ly-glutamyl-2-[2-(2-aminoethoxy)ethoxy]acetyl-2-[2-(2-aminoethoxy)ethoxy]acetyl, pictured herein below

and wherein X2 is attached to the s-amino group of Lys ²⁰ via the amide bond when indicated by Alternatively, the above two peptides, or any fragment thereof, are represented throughout the present application by the use of the “one-letter code” for amino acids/peptides, according to formula I:

HX1EGTFTSDVSSYLEGQAAK(X2)EFIAWLVRGRG (I)

wherein X1 and X2 are as defined above, respectively.

In a further option, the above two peptides, or any fragment thereof, are represented throughout the present application by use of their chemical structure.

Hereinafter, the fragments will also be referred to by their corresponding positions in the aforementioned sequence. For example, a "fragment (1-16)" refers to the peptide sequence consisting of the amino acids present in the above sequence at positions 1 to 16. The preparation of GLP-1 analogs has been described in several patent applications. patent. Both sequential and fragment-based SPPS (solid phase peptide synthesis)-based approaches were followed. It should be noted that the branched structure of liraglutide and semaglutide and their low aqueous solubility drastically impede their synthesis and purification.

In a fragment-based peptide synthesis, the C-terminal carboxylic group of one of the two reacting fragments must be activated prior to coupling. However, activation of the C-terminal carboxylic groups of peptide fragments can lead to the formation of activated ester and oxazolone, which can readily racemize under basic conditions during coupling (NL Benoiton. Chemistry of Peptide Synthesis. Tailor and Francis Group. 2006). Therefore, the crude peptides thus obtained contain impurities corresponding to the isomerized product. Since isomeric impurities have very similar retention times to the required target peptide, it is very difficult to separate these impurities from said peptide. One way to address this problem was to split the peptide sequence of the GLP-1 analogs at Gly ⁴ and/or Gly ¹⁶ positions and assemble the fragments in a final step (as disclosed, for example, in CN104004083 and WO2016046753).

However, the extremely low solubility of the protected fragment (1-16) in the solvents commonly used in peptide synthesis makes final coupling with the schematic fragment (1-16) fragment (17-31) very difficult, especially when the C-terminal fragment (ie, fragment (17-31)) is immobilized on a solid support.

Furthermore, long peptides such as liraglutide and semaglutide are prone to intermolecular aggregation of peptide chains. This problem, along with intramolecular peptide folding, is particularly serious during peptide synthesis, decreasing the efficiency of amino acid couplings and deprotections, leading to the formation of many side products or even the impossibility of obtaining the target peptide. . WO 2017/127007 discloses a process for the preparation of liraglutide, wherein a protected fragment comprising an internal pseudoproline dipeptide is coupled to a protected liraglutide fragment.

Therefore, there remains a need to develop an efficient, simple and industrially viable process for preparing liraglutide and semaglutide. Therefore, an improved process is highly desirable to overcome the synthesis problems not solved by the prior art and to provide both peptides in high yield and with a favorable impurity profile.

Described herein is a new and improved solid-phase synthesis of liraglutide and semaglutide that makes use of a convergent approach involving a final fragment ligation step, in which the N-terminal A fragment has a pseudoproline at the site. C-terminal reagent, according to general scheme 1: fragment A - pseudoproline -COOH NH ?-fragment B - solid support

end peptide

In particular, a strategy is devised in which the protected N-terminal pseudoproline fragment (or fragment A) is prepared by SPPS and then cleaved from the solid support, while maintaining the side chain protections and the a-amino group, and the solid supported C-terminal fragment (or fragment B) is prepared by SPPS or CSPPS (converging SPPS) and only its α-amino group is deprotected prior to ligation of the final fragment. The complete sequenced peptide is finally excised from the solid support and deprotected, and optionally purified to obtain pure liraglutide or semaglutide, respectively. Surprisingly, it was found that the use of pseudoproline at the C-terminus in the present fragment-based approach effectively reduces the formation of impurities so that the obtained crude peptide can be more easily purified to give the final peptide in high overall yield. .

Object of the invention

An object of the present invention is to provide an improved solid phase process for the preparation of ^g LP-1 analogues, or a pharmaceutically acceptable salt thereof, which results in crude peptides with high yield and good performance profile. impurities, facilitating the final purification.

Another object of the present invention is to provide an efficient, simple and industrially feasible process for preparing liraglutide and semaglutide.

A further object of the present invention is to provide a process for preparing liraglutide and semaglutide in higher yield and purity than those achieved in the state of the art.

A further object of the present invention is to provide peptide intermediates useful for the synthesis of liraglutide or semaglutide, or a pharmaceutically acceptable salt thereof.

Summary of the invention

The present invention provides a convergent method for the preparation of some GLP-1 analogues in solid phase synthesis, comprising the final step of coupling a first N-terminal protected peptide fragment characterized by having a pseudoproline at the C-terminal reactive site. (N-terminal protected pseudoproline fragment A), with a second C-terminal protected peptide fragment (fragment B) attached to a solid support.

Fragment A has a C-terminal pseudoproline reactive site and is protected at its N-terminal α-amino group. Fragment A is preferably protected on its amino acid side chains.

Fragment A is selected from the group consisting of:

His-X1-Glu-Gly-Thr-Phe-Thr-Ser(y z'zpro),

called fragment A (1-8) or SEQ ID NO: 3,

His-X1-Glu-Gly-Thr-Phe-Thr-Ser-Asp-Val-Ser(^ zzpro),

referred to as fragment A(1-11) or SEQ ID NO: 4,

His-X1-Glu-Gly-Thr-Phe-Thr-Ser-Asp-Val-Ser-Ser(^ zzpro),

referred to as fragment A(1-12) or SEQ ID NO: 5,

His-X1-Glu-Gly-Thr-Phe-Thr(^ zzpro),

referred to as fragment A(1-7) or SEQ ID NO: 6,

and

His-Xi-Glu-Gly-Thr(^z'zpro), referred to as fragment A(1-5) or SEQ ID NO: 7,

where X ¹ is Ala for liraglutide and Aib for semaglutide, and

wherein aa(^zzpro) indicates that the C-terminal amino acid (aa), preferably Ser or Thr, is protected with pseudoproline, according to the following formula:

in which

Y is hydrogen for Ser and methyl (Me) for Thr,

Z is hydrogen or methyl (Me), and

R ¹ is the side chain of the amino acid next to the pseudoproline protected amino acid.

The fragment number annotation in parentheses, eg, (1-8), simply indicates the position of the fragment with respect to the complete peptide sequence.

Therefore, for example, aa-Ser(^Me'Mepro) indicates:

where R ¹ is the side chain of aa, and

Thr-Ser(^zzpro) in fragment A(1-8) as defined above indicates:

where Z is as defined above.

Stereochemistry is not defined in the above structures for simplicity. L-amino acids are always used unless otherwise indicated.

Fragment B is attached to a resin, or solid support, at its C-terminal amino acid, ie, glycine 31 (Gly31). The solid support is preferably selected from Wang's resin, 2-chlorotrityl chloride (CTC) resin, trityl chloride resin, mBh resin and 4-MeO-MBH. The solid support is more preferably selected from Wang's resin, 2-chlorotrityl chloride (CTC) resin and trityl chloride resin. Fragment B is preferably protected on its amino acid side chains, including the X ² residue of Lys20.

Fragment B is selected from the group consisting of

Asp-Val-Ser-Ser-Try-Leu-Glu-Gly-Gln-Ala-Ala-Lys(X2)-Glu-Phe-Ile-Ala-Trp-Leu-Val-Arg-Gly-Arg-Gly-resin , referred to as fragment B(9-31) or SEQ ID NO: 8,

Ser-Tyr-Leu-Glu-Gly-Gln-Ala-Ala-Lys(X2)-Glu-Phe-Ile-Ala-Trp-Leu-Val-Arg-Gly-Arg-Gly-resin, called fragment B (12 -31) or SEQ ID NO: 9,

Tyr-Leu-Glu-Gly-Gln-Ala-Ala-Lys(X2)-Glu-Phe-Ile-Ala-Trp-Leu-Val-Arg-Gly-Arg-Gly-resin, called fragment B (13-31 ) or SEQ ID NO: 10,

Ser-Asp-Val-Ser-Ser-Tyr-Leu-Glu-Gly-Gln-Ala-Ala-Lys(X2)-Glu-Phe-Ile-Ala-Trp-Leu-Val-Arg-Gly-Arg-Gly -resin, designated fragment B(8-31) or s Eq ID NO: 11,

and

Phe-Thr-Ser-Asp-Val-Ser-Ser-Tyr-Leu-Glu-Gly-Gln-Ala-Ala-Lys(X2)-Glu-Phe-Ile-Ala-Trp-Leu-Val-Arg-Gly -Arg-Glyresin, designated fragment B(6-31) or SeQ ID NO: 12,

in which

X2 is N-(1-oxohexadecyl)-L-y-glutamyl (Pal-Glu) for liraglutide, and

X2 is N-(17-carboxy-1-oxoheptadecyl)-L-y-glutamyl-2-[2-(2-aminoethoxy)ethoxy]acetyl-2-[2-(2-aminoethoxy)ethoxy]acetyl to semaglutide, optionally protected .

Fragment A and fragment B are ligated so that the complete sequence of either liraglutide or semaglutide is obtained. The obtained 30 aa peptide, preferably with protected side chain, is then deprotected and/or cleaved from the solid support to obtain the crude peptide. Such crude peptide is optionally purified to obtain pure liraglutide or semaglutide. The method of the present invention provides liraglutide and semaglutide in high yield and high purity.

Furthermore, the present invention provides intermediate fragments A and B and methods for their preparation.

Detailed description of the invention

The present invention provides a new method for preparing a GLP-1 analogue in high yield and purity. In particular, it consists of a convergent approach on the solid phase. More particularly, the final coupling involves a peptide fragment (fragment A) characterized as having a pseudoproline at the C-terminal reactive site.

Pseudoprolines were developed by Mutter in 1992 (T. Haack, M. Mutter, Serine derived oxazolidines as secondary structure disrupting, solubilizing building blocks in peptide synthesis, Tetrahedron Lett. 1992, 33, 1589-1592). Surprisingly, it has been found that the use of pseudoproline at the C-terminus in the present fragment-based approach effectively reduces the amount of impurities formed, leading to the improved process that is the subject of the present invention. Such a method substantially reduces the formation of impurities so that the crude peptide obtained can be easily purified to give the final peptide in high overall yield.

In a further aspect, the present invention provides the requisite A and B peptide fragments and the method of preparing them. Such synthesis is carried out by conventional SPPS or by CSPPs, both through the use of a coupling reagent system.

The term "peptide fragment" or "fragment" describes a peptide with a partial amino acid sequence, with reference to the sequence of liraglutide or semaglutide. It can optionally be attached to a resin at its C-terminal amino acid. A peptide fragment may be protected or unprotected. A peptide fragment may have a pseudoproline at its C-terminus or at a position internal to the amino acid sequence.

The term "protected peptide fragment" or "protected fragment" describes a peptide fragment that can independently carry protecting groups on its amino acid side chains, or side groups, and/or on its a-amino group.

A fragment, or peptide fragment, can also be indicated with a specific amino acid sequence, such as aa1-aa2-...-aan

in which aaX is the three-letter code of the amino acid in position x, and in which the presence or absence of protecting groups is not defined, either in the side chains or in the a-amino group. The need to protect amino acid side chains during peptide synthesis under certain conditions is known in the art and various strategies have been described to identify suitable protection strategies.

If in this description for a specific amino acid sequence (fragment) side chain protecting groups are specifically mentioned in parentheses accompanying the three-letter amino acid code, it is understood that the remaining amino acids that are represented by a three-letter code simple are checked out.

When the fragment is indicated with

aa1-aa2-...-aan-OH

the C-terminal amino acid aan is intended to have a free carboxylic acid.

When the fragment is indicated with

H-aa1-aa2-...-aan

the N-terminal aa1 amino acid is intended to have a free a-amino group.

When the fragment is indicated with

aa1-aa2-...-aan(^z,zpro)-OH

the C-terminal amino acid aan is intended to be protected with pseudoproline and have a free carboxylic acid, according to the following structure:

where R ¹ is the side chain of aa(n-1), and aan(^zzpro)-OH can be

Ser(^ zzpro)-OH, where Y is hydrogen, and Z is hydrogen or Me, or

Thr(^zzpro)-OH, where Y is Me, and Z is hydrogen or Me.

Pseudoprolines, as α-amino protected dipeptides, are commercially available. For example, the following pseudoproline dipeptides are used in carrying out the present invention:

P-Thr(fBu)-ser(v zzpro)8-OH

P-Val-Ser(^zzpro)11-OH

P-Ser-Ser(^zzpro)12-OH

P-Phe-Thr(^zzpro)7-OH

P-Gly-Thr(^ zzpro)5-OH

wherein P is an α-amino protecting group, and Z is as defined above.

Preferably, the pseudoproline dipeptides used to carry out the present invention are selected from the group consisting of:

Fmoc-Thr(fBu)-Ser(VMe'Mepro)8-OH

Fmoc-Thr(tBu)-Ser(and HHpro)8-OH

Fmoc-Val-Ser(and Me'Mepro)11-OH

Fmoc-Val-Ser(^HHpro)11-OH

Fmoc-Ser(tBu)-Ser(and MeMepro)12-OH

Fmoc-Ser(tBu)-Ser(and HHpro)12-OH

Fmoc-Phe-Thr(and MeMepro)7-OH

Fmoc-Phe-Thr(and HHpro)7-OH

Fmoc-Gly-Thr(^MeMepro)5-OH and

Fmoc-Gly-Thr(and HHpro)5-OH

The introduction of pseudoprolines into a peptide sequence can be carried out on a solid phase under standard coupling conditions. Once the peptide is cleaved from the resin by acidolysis, the pseudoproline is hydrolyzed, yielding the two corresponding native amino acids.

The term "resin" or "solid support" describes a functionalized insoluble polymer to which an amino acid or peptide fragment can be attached and which is suitable for the elongation of amino acids until the complete desired sequence is obtained.

SPPS can be defined as a procedure in which a peptide anchored by its C-terminal amino acid to a resin is assembled by sequential addition of the optionally protected amino acids that constitute its sequence. It comprises loading a first amino acid, or peptide, or dipeptide of pseudoproline with α-amino protected as defined above, on a resin and is followed by the repetition of a sequence of steps called the cycle, or elongation step, consisting of the cleavage of the α-amino protecting group and subsequent protected amino acid attachment.

The formation of a peptide bond between two amino acids, or between an amino acid and a peptide fragment, or between two peptide fragments, may involve two steps. First, the activation of the free carboxyl group for a time ranging from 5 minutes to 2 hours, then the nucleophilic attack of the amino group on the activated carboxylic group.

The cycle can be repeated sequentially until the desired peptide sequence is achieved.

Finally, the peptide is deprotected and/or cleaved from the resin.

As a reference for SPPS, see, for example, Knud J. Jensen et al. (eds.), Peptide Synthesis and Applications, Methods in Molecular Biology, vol. 1047, Springer Science, 2013.

In a preferred aspect of the present invention, an acid sensitive resin is used. More preferably, it is selected from Wang resin, 2-chlorotrityl chloride (CTC) resin, trityl chloride resin, MBH resin and 4-MeO-BH. More preferably, the resin is selected from Wang resin, 2-chlorotrityl chloride (CTC) resin and trityl chloride resin.

Even more preferably, the resins of Wang and MBH are used for the preparation of fragment B and are therefore involved in the coupling of fragment A to fragment B. Even more preferably, the resin of Wang is used for the preparation of fragment B. fragment B and is therefore involved in the coupling of fragment A with fragment B.

Preferably, CTC resin is used for the preparation of fragment A, and fragment A is cleaved from the resin prior to coupling with fragment B.

In a preferred aspect of the present invention, loading of the first C-terminal amino acid into the Wang resin is carried out after swelling the resin in a suitable solvent, preferably DMF, filtering the resin, and adding the resin solution to the resin. amino acid protected with an activating agent, such as a carbodiimide, in the presence of a base, such as dimethylaminopyridine (DMAP). In case of using a trityl chloride resin, and in particular the CTC resin, after swelling the resin in a suitable solvent, preferably DCM, and filtering the resin, it is add a solution of the amino acid protected with an organic base, preferably diisopropylethylamine (DIEA). In a further preferred aspect of the present invention, for the preparation of fragment A, the first step is the loading of the appropriate pseudoproline dipeptide onto the CTC resin, after swelling the resin in a suitable solvent, preferably DCM.

In a preferred aspect of the present invention, after loading the first C-terminal amino acid or pseudoproline dipeptide onto the resin, an additional step is performed to block unreacted sites to prevent truncated sequences and any side reactions. Such a stage is often referred to as "capping."

Capping is accomplished by a brief treatment of the loaded resin with a large excess of a highly reactive unhindered reagent, which is chosen based on the unreacted sites to be capped. When a Wang resin is used, the unreacted sites are hydroxyl groups, which are preferably capped by treatment with an acid derivative, such as an anhydride, in a basic medium, for example, with a DMF/DIEA/Ac2O mixture, preferably with a 17/2/1 v/v/v composition or with a 5/2/1 v/v/v composition, or with a 10% Ac2O solution in DMF.

When using a CTC resin, the unreacted sites are chloros, which are preferably capped by treatment with an alcohol in a basic medium, for example with a DCM/DIEA/MeOH mixture, preferably with a 17/2/ 1 v/v/v. Then, after washing with DCM, the resin is further treated with DCM/DIEA/Ac2O mixture, preferably with a 17/2/1 v/v/v composition, to cap hydroxyl groups possibly resulting from chlorine hydrolysis.

As an alternative to loading the first C-terminal amino acid, preloaded resins are used in the preparation of peptide fragments. These are commercially available Wang/CTC resins with attached Fmoc-protected L- or D-amino acids. Therefore, for example, Wang's Fmoc-Gly resin is preferably used for the synthesis of fragment B, and Fmoc-pseudoproline resin can be used for the synthesis of fragment A. In a preferred aspect of the present invention, charging of the The first C-terminal amino acid on the resin is determined spectrophotometrically, as described, for example, in Knud J. Jensen et al. (eds.), Peptide Synthesis and Applications, Methods in Molecular Biology, vol. 1047, Springer Science, 2013.

In a preferred aspect of the present invention, each amino acid may be protected at its α-amino group and/or at its side chain functional groups.

Preferably, the protecting group of the α-amino groups of the amino acids that is used for the SPPS is of the carbamate type. The most preferred protecting groups are the 9-fluorenylmethoxycarbonyl (Fmoc) group, which can be removed under basic conditions, and the tert-butyloxycarbonyl (Boc) group, which can be removed under acidic conditions.

The amino acid side chain functional groups are optionally protected with groups which are generally stable during the coupling steps and during removal of the α -amino protecting group, and which, in turn, can be removed under suitable conditions. Such suitable conditions are generally the opposite of the conditions under which the α-amino groups are deprotected. The protecting groups of the amino acid side chain functional groups used in the present disclosure can generally be removed under acidic conditions as opposed to the basic conditions generally used to deprotect Fmoc protecting groups.

In a preferred aspect of the present invention, such side chain protecting groups are specified by individual amino acid occurring in the relevant GLP-1 analogue sequence, as follows:

the arginine (Arg)guanidinium group is preferably protected by a protecting group selected from the group consisting of methoxy-2,3,6-trimethylbenzenesulfonyl (Mtr), 2,2,5,7,8-pentamethylchromanyl-6-sulfonyl (Pmc ) and 2,2,4,6,7-pentamethyl-dihydrobenzofuran-5-sulfonyl (Pbf), all eliminable under acidic conditions; more preferably, the Pbf group is used;

the hydroxyl group of serine (Ser), threonine (Thr) or tyrosine (Tyr) is preferably protected by a group selected from the group consisting of trityl (Trt), tert-butyldimethylsilyl (TBDMS) and tert-butyl (tBu); more preferably, the group tBu is used;

the carboxylic group of glutamic (Glu) or aspartic (Asp) acid is preferably protected by a group selected from the group consisting of 2-phenylisopropyl (2-Phi-Pr), tBu ester, benzyl ester (OBzl), allyl ester (OAII ); more preferably, the tBu ester is used;

the indole nitrogen of tryptophan (Trp) is preferably protected by a group selected from the group consisting of tert-butyloxycarbonyl (Boc), formyl (For); more preferably, the tert-butyloxycarbonyl (Boc) group is used;

the histidine imidazole nitrogen (His) is preferably protected by a group selected from the group consisting of ferrc-butyloxycarbonyl (Boc), monomethoxytrityl (Mmt), 3-methoxybenzyloxymethyl (MBom), 2-chlorotrityl (Clt), trityl (Trt ); more preferably, the trityl (Trt) group is used.

Commercially available protected L-amino acids are generally used. As for the introduction of the lipophilic albumin-binding moiety X ² into Lys20, the commercially available Fmoc-Lys(Pal-Glu-OfBu)-OH depicted here below is used in the preparation of liraglutide:

In the preparation of semaglutide, the protected lysine derivative depicted here below is used:

Such a derivative is commercially available, and its preparation is also described, for example, in CN104356224. Such a derivative is also indicated as Fmoc-Lys[(fBuOOC-(CH ² ) ¹⁶ -CO-yGlu-OfBu)-AEEA-AEEA]-OH, where AEEA represents 2-[2-(2-aminoethoxy)ethoxy] acetyl.

The tBu ester protecting groups of the X ² moiety building blocks are all removable under acidic conditions and are therefore cleaved at the end of the preparation process.

In a preferred aspect of the present invention, the coupling steps are performed in the presence of a coupling reagent.

Preferably, the coupling reagent is selected from the group consisting of N-hydroxysuccinimide (NHS), N,N'-diisopropylcarbodiimide (DIC), N,N'-dicyclohexylcarbodiimide (DCC), (benzotriazol-1-yloxy)tripyrrolidinophosphonium hexafluorophosphate (PyBOP), 2-(7-aza-1H-benzotriazol-1-yl)-1,1,3,3-tetramethyluronium hexafluorophosphate (HATU), 2-(1H-benzotriazol-1-yl)-1 hexafluorophosphate ,1,3,3-tetramethyluronium (HBTU) and ethyldimethylaminopropylcarbodiimide (EDC). More preferably, the reaction is carried out in the presence of N,N'-diisopropylcarbodiimide.

In a preferred aspect of the present invention, the coupling steps are also carried out in the presence of an additive. The presence of an additive, when used in the coupling reaction, reduces configuration loss at the carboxylic acid residue, increases coupling rates, and reduces the risk of racemization.

Preferably, the additive is selected from the group consisting of 1-hydroxybenzotriazole (HOBt), 2-hydroxypyridine N-oxide, N-hydroxysuccinimide (NHS), 1-hydroxy-7-azabenzotriazole (HOAt), endo-N-hydroxy- 5-norbornene-2,3-dicarboxamide, 5-(hydroxyimino)1,3-dimethylpyrimidine-2,4,6-(1H,3H,5H)-trione (Oxyma-B) and 2-cyano-2-hydroxyiminoacetate ethyl (OxymaPure). More preferably, the additive is selected from the group consisting of 1-hydroxybenzotriazole (HOBt), 2-hydroxypyridine N-oxide, N-hydroxysuccinimide (NHS), 1-hydroxy-7-azabenzotriazole (HOAt), endo-N-hydroxy -5-norbornene-2,3-dicarboxamide, and ethyl 2-cyano-2-hydroxyimino-acetate (OxymaPure). Even more preferably, the reaction is carried out in the presence of 2-cyano-2-hydroxyimino-acetate.

In a preferred aspect, the coupling steps are carried out in the presence of a detergent. Preferred detergents for fragment coupling according to the described synthesis are nonionic detergents, for example, Triton X-100 (also called TX-100 or polyethylene glycol tert-octylphenyl ether) or Tween 20, most preferably Triton X-100 . For example, TX-100 is used as a 1% solution in DMF:DCM 50:50 v/v.

In a preferred aspect of the present invention, the coupling steps are performed in a ^solvent selected from the group consisting of DMF, DCM, ^tHf , NMP, DMA, or mixtures thereof. More preferably, the coupling is carried out in DMF when using a Wang resin, and is carried out in DCM when using a CTC resin.

In a preferred aspect of the present invention, the coupling steps are carried out at a temperature that can vary in the range of 10-70°C. Preferably the temperature ranges from room temperature (ie 20°C) to 50°C, more preferably the temperature ranges from 35-45°C, even more preferably the coupling step is carried out out at 40°C.

In a preferred aspect of the present invention, the α-amino protecting groups are cleaved under basic conditions. In particular, the Fmoc protecting group is cleaved by treatment with a suitable secondary amine selected from the group consisting of piperidine, pyrrolidine, piperazine and DBU, preferably piperidine. More preferably, Fmoc deprotection is carried out using a 20% solution of piperidine in DMF.

In another preferred aspect of the present invention, the α-amino protecting groups are cleaved under acidic conditions. In particular, the Boc protecting group is cleaved by treatment with a strong acid solution, for example with trifluoroacetic acid (TFA), hydrochloric acid (HCl) or phosphoric acid (H3PO4). More preferably, Boc deprotection is carried out during cleavage of the peptide or peptide fragment from the resin by treatment with a TFA-based mixture, which is selected according to the type of resin. For example, a mixture of TFA/water/phenol/TIS (v/v/v/v 88/5/5/2) is used for a Wang resin, where TIS is a scavenger, as also described next.

In a preferred aspect of the present invention, once the desired final peptide or peptide fragment has been obtained according to SPPS or CSPPS as described above and is attached to its solid support, final deprotection and/or cleavage of such is performed. solid support. Preferably, such a step is carried out using a specific mixture individualized for the resin used, under acidic or slightly acidic conditions.

When using a CTC resin, the cleavage step is preferably performed by treatment using a mixture of HFIP:DCM (30:70 v/v) or 1-2% v/v TFA solution in DCM. In particular, when a CTC resin is used in the preparation of fragment A or another fragment, such cleavage does not remove the α-amino protecting group or side chain protecting groups, thus giving a fully protected, ready fragment. to react into its free C-terminal carboxylic acid.

When using a Wang resin, treatment with a cleavage mixture, comprising TFA and scavengers, provides for both side chain deprotection, including deprotection of the X2 moiety, and cleavage from the resin. In particular, when a Wang resin is used in the preparation of fragment B, the final cleavage produces the crude GLP-1 analog. Scavengers are substances, such as, for example, anisole, thioanisole, triisopropylsilane (TIS), 1,2-ethanedithiol, and phenol, that can minimize the modification or destruction of sensitive unprotected side chains, such as those of arginine residues. , in the cleavage environment. Such a cleavage/deprotection step is preferably performed using a TFA/thioanisole/anisole/dodecanethiol mixture, for example with a composition v/v/v/v 90/5/2/3, or a TFA/water/ phenol/TIS, for example, with v/v/v/v 88/5/5/2.

In a preferred aspect of the present invention, the crude peptide obtained by cleavage from the resin, whether it is a final GLP-1 fragment or analog, is purified to increase its purity.

For this purpose, a solution of the peptide is loaded onto an HPLC column with a suitable stationary phase, preferably C18 or C8 modified silica, and an aqueous mobile phase comprising an organic solvent, preferably acetonitrile or methanol. A mobile phase gradient is applied, if necessary. Peptide with the desired purity is collected and optionally lyophilized.

Accordingly, the present invention provides a convergent method for the preparation of liraglutide or semaglutide in solid phase synthesis, wherein the method comprises final coupling of at least one N-terminal protected pseudoproline fragment A with a C-fragment B. capped end, followed by simultaneous or sequential deprotection and cleavage from the solid support, and optionally final purification by chromatography.

The present invention further provides intermediate fragments A and B and methods for their preparation.

In one aspect, fragment A is an N-terminal fragment of the desired GLP-1 analog peptide, having a pseudoproline at the C-terminal reactive site. In a preferred aspect, fragment A is protected at its a-amino group, even more preferably with a Boc group. In another preferred aspect, fragment A is protected at its side chain functional groups with suitable protecting groups.

Fragment A is selected from the group consisting of:

His-Xi-Glu-Gly-Thr-Phe-Thr-Ser(vzzpro), called fragment A (1-8),

His-X ¹ -Glu-Gly-Thr-Phe-Thr-Ser-Asp-Val-Ser(^z'zpro), designated fragment A (1-11),

His-X ¹ -Glu-Gly-Thr-Phe-Thr-Ser-Asp-Val-Ser-Ser(^zzpro), called fragment A (1-12),

His-X ¹ -Glu-Gly-Thr-Phe-Thr(^zzpro), called fragment A (1-7),

and

His-X ¹ -Glu-Gly-Thr(^zzpro), called fragment A (1-5)

where X ¹ is Ala for liraglutide and Aib for semaglutide, and

wherein (^zzpro) indicates the pseudoproline used, and has the meaning as defined above. Preferably, fragment A is selected from the group consisting of

Boc-His-X ¹ -Glu-Gly-Thr-Phe-Thr-Ser(^MeMepro),

Boc-His-X ¹ -Glu-Gly-Thr-Phe-Thr-Ser-Asp-Val-Ser(^MeMepro),

Boc-His-X ¹ -Glu-Gly-Thr-Phe-Thr-Ser-Asp-Val-Ser-Ser(^MeMepro),

Boc-His-X ¹ -Glu-Gly-Thr-Phe-Thr(^MeMepro)

and

Boc-His-X ¹ -Glu-Gly-Thr(yMe,Mepro),

where X ¹ y (^MeMepro) are as defined above.

More preferably, for the preparation of liraglutide, fragment A is selected from the group consisting of: Boc-His(Trt)1-Ala-Glu(OtBu)-Gly-Thr(tBu)-Phe-Thr(tBu)-Ser (yMeMepro)8-OH,

Boc-His(Trt)1-Ala-Glu(OfBu)-Gly-Thr(f-Bu)-Phe-Thr(f-Bu)-Ser(f-Bu)-Asp(Of'Bu)-Val-Ser (njMe'Mepro)11-OH,

Boc-His(Trt)1-Ala-Glu(OfBu)-Gly-Thr(f-Bu)-Phe-Thr(f-Bu)-Ser(f-Bu)-Asp(Of'Bu)-Val-Ser (f-Bu)-Ser(njMe'Mepro)12-OH,

Boc-His(Trt)1-Ala-Glu(OfBu)-Gly-Thr(fBu)-Phe-Thr(^MeMepro)7-OH,

Boc-His(Trt)1-Ala-Glu(OtBu)-Gly-Thr(^MeMepro)5-OH,

Boc-His(Trt)1-Aib-Glu(OfBu)-Gly-Thr(fBu)-Phe-Thr(fBu)-Ser(yMeMepro)8-OH,

Boc-His(Trt)1-Aib-Glu(OtBu)-Gly-Thr(f-Bu)-Phe-Thr(f-Bu)-Ser(f-Bu)-Asp(Of'Bu)-Val-Ser (njMe'Mepro)11-OH,

Boc-His(Trt)1-Aib-Glu(Oí'Bu)-Gly-Thr(í'Bu)-Phe-Thr(í'Bu)-Ser(í'Bu)-Asp(OfBu)-\/al -Ser(í'Bu)-Ser(njMe'Mepro)12-OH,

Boc-His(Trt)1-Aib-Glu(OfBu)-Gly-Thr(fBu)-Phe-Thr(^MeMepro)7-OH, and

Boc-His(Trt)1-Aib-Glu(OfBu)-Gly-Thr(^MeMepro)5-OH.

Even more preferably, fragment A is

Boc-His(Trt)1-Ala-Glu(OfBu)-Gly-Thr(fBu)-Phe-Thr(fBu)-Ser(^ Me'Mepro)8-OH, also called fragment 4, which is a specified form from fragment A (1-8).

In another embodiment, where the preparation of semaglutide is intended, fragment A is preferably Boc-His-Aib-Glu-Gly-Thr-Phe-Thr-Ser(and MeMepro)

and even more preferably

Boc-His(Trt)1-Aib-Glu(OfBu)-Gly-Thr(fBu)-Phe-Thr(tBu)-Ser(^MeMepro)8-OH, also called fragment 7, which is a specified form of the fragment A(1-8).

Fragment A is prepared either by stepwise SPPS or by LPPS. For example, fragment A is prepared on a solid phase by starting with loading onto a resin the appropriate C-terminal pseudoproline dipeptide, eg, Fmoc-Thr(tBu)-Ser(^MeMepro)8-OH, and subsequent elongation. of the peptide chain coupling the amino acids of sequence one by one, alternating stages of coupling and deprotection of the a-amino group. Preferably, stepwise Fmoc SPPS on a CTC resin is used in the preparation of fragment A, and the synthesis is completed using a Boc-protected last amino acid, ie, Boc-His(Trt)-OH. Upon completion of its sequencing, fragment A is excised from the solid support while retaining both the α-amino group and side chain protecting groups.

As examples for the preparation of fragment A, the preparation of fragment 4 and fragment 7 are described in the examples of the present disclosure.

Similarly, all of the A fragments defined above are prepared in a similar manner.

In another embodiment, fragment B is a C-terminal fragment of the desired GLP-1 analog peptide, attached to a solid support, which is preferably a Wang resin. In other preferred embodiments, the solid support is a MBH resin. .

Fragment B is selected from the group consisting of

Asp-Val-Ser-Ser-Tyr-Leu-Glu-Gly-Gln-Ala-Ala-Lys(X ² )-Glu-Phe-Ile-Ala-Trp-Leu-Val-Arg-Gly-Arg-Gly- wang Resin,

Ser-Tyr-Leu-Glu-Gly-Gln-Ala-Ala-Lys(X ² )-Glu-Phe-Ile-Ala-Trp-Leu-Val-Arg-Gly-Arg-Gly-Wang's resin,

Tyr-Leu-Glu-Gly-Gln-Ala-Ala-Lys(X ² )-Glu-Phe-Ile-Ala-Trp-Leu-Val-Arg-Gly-Arg-Gly-Wang's resin,

Ser-Asp-Val-Ser-Ser-Tyr-Leu-Glu-Gly-Gln-Ala-Ala-Lys(X ² )-Glu-Phe-Ile-Ala-Trp-Leu--Val-Arg-Gly-Arg -Wang's Gly-resin,

and

Phe-Thr-Ser-Asp-Val-Ser-Ser-Tyr-Leu-Glu-Gly-Gln-Ala-Ala-Lys(X ² )-Glu-Phe-Ile-Ala-Trp-Leu-Val-Arg- Gly-Arg-Gly- resin of Wang, in which

X2 is N-(1-oxohexadecyl)-L-y-glutamyl (Pal-Glu) for the synthesis of liraglutide, and

X2 is N-(17-carboxy-1-oxoheptadecyl)-L-y-glutamyl-2-[2-(2-aminoethoxy)ethoxy]acetyl-2-[2-(2-aminoethoxy)ethoxy]acetyl for the synthesis of semaglutide .

Preferably, for the preparation of liraglutide, fragment B is selected from the group consisting of H-Asp(OtBu)9-Val-Ser(tBu)-Ser(fBu)-Tyr(tBu)-Leu-Glu(OtBu)-Gly16-Gln(Trt)-Ala-Ala-Lys(Pal-Glu-OtBu )-Glu(OfBu)-Phe-Ile-Ala-Trp(NínBoc)-Leu-Val-Arg(Pbf)-Gly-Arg(Pbf)-Gly31-res¡na de Wang,

also called fragment 3,

H-Ser(tBu)-Tyr(íBu)-Leu-Glu(OtBu)-Gly16-Gln(Trt)-Ala-Ala-Lys(Pal-Glu-OfBu)-Glu(OíBu)-Phe-Ile-Ala- Trp([\rBoc)-Leu-Val-Arg(Pbf)-Gly-Arg(Pbf)-Gly31-resin of Wang,

H-Tyr(fBu)-Leu-Glu(OtBu)-Gly16-Gln(Trt)-Ala-Ala-Lys(Pal-Glu-OtBu)-Glu(OtBu)-Phe-Ile-Ala-Trp(N,nBoc )-Leu-Val-Arg(Pbf)-Gly-Arg(Pbf)-Gly31-Wang resin, H-Ser(tBu)-Asp(OtBu)9-Val-Ser(tBu)-Ser(íBu) -Tyr(tBu)-Leu-Glu(OtBu)-Gly16-Gln(Trt)-Ala-Ala-Lys(Pal-Glu-OfBu)-Glu(OtBu)-Phe-Ile-Ala-Trp(N/í' Boc)-Leu-Val-Arg(Pbf)-Gly-Arg(Pbf)-Gly31-Wang's resin

and

H-Phe-Thr(fBu)-Ser(tBu)-Asp(0fBu)9-Val-Ser(íBu)-Ser(tBu)-Tyr(fBu)-Leu-G lu(0tBu)-Gly16-Gln( Trt)-Ala-Ala-Lys(Pal-Glu-OtBu)-Glu(OtBu)-Phe-Ile-Ala-Trp(N"IBoc)-Leu-Val-Arg(Pbf)-Gly-Arg( Pbf)- Gly31-resin from Wang.

Even more preferably, for the liraglutide preparation, fragment B is:

H-Asp(OfBu)9-Val-Ser(fBu)-Ser(fBu)-Tyr(fBu)-Leu-Glu(OtBu)-Gly16-Gln(Trt)-Ala-Ala-Lys(Pal-Glu-OfBu )-Glu(OfBu)-Phe-Ile-Ala-Trp(N'nBoc)-Leu-Val-Arg(Pbf)-Gly-Arg(Pbf)-Gly31-Wang resin,

also called fragment 3, which is a specified form of fragment B (9-31).

In another embodiment where preparation of semaglutide is intended, fragment B is preferably Asp-Val-Ser-Ser-Tyr-Leu-Glu-Gly-Gln-Ala-Ala-Lys[(ROOC-(CH2)i6- CO-yGlu-OR0-AEEA-AEEA]-Glu-Phe-Ile-Ala-Trp-Leu-Val-Arg-Gly-Arg-Gly-resin,

wherein R and R', the same or different, are carboxylic acid protecting groups, preferably esters; more preferably fragment B is

H-Asp(OfBu) ⁹ -Val-Ser(fBu)-Ser(fBu)-Tyr(fBu)-Leu-Glu(OtBu)-Gly-Gln(Trt)-Ala-Ala-Lys[(fBuOOC-(CH2 )16-CO-yGlu-OfBu)-AEEA-AEEA]-Glu-(OfBu)-Phe-Ile-Ala-Trp(NmBoc)-Leu-Val-Arg(Pbf)-Gly-Arg(Pbf)-Gly ³¹ -MBH resin, also called fragment 6, which is a specified form of fragment B (9-31).

Fragment B is prepared either by stepwise SPPS or by CSPPS, or by LPPS. For example, fragment B is prepared in solid phase starting with the loading on the resin of the protected C-terminal amino acid, that is to say Gly ³¹ , and the subsequent elongation of the peptide chain by coupling the sequence amino acids one by one, alternating steps of coupling and deprotection of a -amino group. Alternatively, fragment B is prepared by CSPPS by coupling two or more peptide fragments.

In a preferred embodiment of the present invention, the preparation of fragment B comprises coupling the appropriate N-terminal fragment C to the C-terminal fragment D, according to scheme 2:

fragment C-COOH NF^-fragment D-solid support

NH^-fragment B - solid support

Fragment D is attached to a solid support via Gly ³¹ and is preferably protected on its amino acid side chains. In a preferred embodiment of the present invention, fragment D is

Gln ¹⁷ -Ala-Ala-Lys(X2)-Glu-Phe-Ile-Ala-Trp-Leu-Val-Arg-Gly-Arg-Gly ³¹ -resin, also called fragment D(17 31) or SEQ ID NO: 13.

Fragment C is a suitable fragment selected from the group consisting of:

Asp-Val-Ser-Ser-Tyr-Leu-Glu-Gly, designated fragment C(9-16) or SEQ ID NO: 14,

Ser-Tyr-Leu-Glu-Gly, designated fragment C(12-16) or SEQ ID NO: 15,

Tyr-Leu-Glu-Gly, designated fragment C(13-16) or SEQ ID NO: 16,

Ser-Asp-Val-Ser-Ser-Tyr-Leu-Glu-Gly, designated fragment C(8-16) or SEQ ID NO: 17,

and

Phe-Thr-Ser-Asp-Val-Ser-Ser-Tyr-Leu-Glu-Gly, designated fragment C(6-16) or SEQ ID NO: 18.

Fragment C has a C-terminal glycine which is conveniently reacted to form the required amide bond under the conditions discussed above. Fragment C is also preferably protected at its amino acid side chains and at its a-amino group.

Fragment C is preferably selected from the group consisting of:

Fmoc-Asp(OfBu) ⁹ -Val-Ser(fBu)-Ser(fBu)-Tyr(fBu)-Leu-Glu(OtBu)-Gly ¹⁶ -OH,

Fmoc-Ser(fBu)-Tyr(fBu)-Leu-Glu(OtBu)-Gly ¹⁶ -OH,

Fmoc-Tyr(fBu)-Leu-Glu(OtBu)-Gly ¹⁶ -OH,

Fmoc-Ser(fBu)-Asp(OfBu) ⁹ -Val-Ser(fBu)-Ser(fBu)-Tyr(fBu)-Leu-Glu(OtBu)-Gly ¹⁶ -OH

and

Fmoc-Phe-Thr(fBu)-ser(fBu)-Asp(OfBu) ⁹ -Val-ser(fBu)-ser(fBu)-Tyr(fBu)-Leu-Glu(OtBu)-Gly ¹⁶ -OH.

Most preferably, fragment C is

Fmoc-Asp(OfBu) ⁹ -Val-Ser(fBu)-Ser(fBu)-Tyr(fBu)-Leu-Glu(OtBu)-Gly ¹⁶ -OH, also called fragment 2.

Such fragment C is conveniently used for the preparation of both liraglutide and semaglutide, as will be described in the examples of the present disclosure.

In a more preferred embodiment of the present invention

Asp-Val-Ser-Ser-Tyr-Leu-Glu-Gly-Gln-Ala-Ala-Lys(X2)-Glu-Phe-Ile-Ala-Trp-Leu-Val-Arg-Gly-Arg-Gly-resin of Wang is obtained by a fragment C (9-16) fragment D (17-31) docking strategy, preferably docking

Asp ⁹ -Val-Ser-Ser-Tyr-Leu-Glu-Gly ¹⁶ -OH

with

H-Gln-Ala-Ala-Lys(X2)-Glu-Phe-Ile-Ala-Trp-Leu-Val-Arg-Gly-Arg-Gly ³¹ resin

in which

X2 is N-(1-oxohexadecyl)-L-y-glutamyl for the synthesis of liraglutide, and

X2 is N-(17-carboxy-1-oxoheptadecyl)-L-y-glutamyl-2-[2-(2-aminoethoxy)ethoxy]acetyl-2-[2-(2-aminoethoxy)ethoxy]acetyl for the synthesis of semaglutide ;

and wherein the resin is preferably a Wang resin.

In an even more preferred embodiment, if preparation of liraglutide is intended, the preferred fragment 3 is obtained (ie, H-Asp(OfBu) ⁹ -Val-Ser(fBu)-Ser(fBu)-Tyr(fBu)- Leu-Glu(OtBu)-Gly ¹⁶ -Gln(Trt)-Ala-Ala-Lys(Pal-Glu-OfBu)-Glu(OfBu)-Phe-Ile-Ala-Trp(N'"Boc)-Leu-Val -Arg(Pbf)-Gly-Arg(Pbf)-Gly ³¹ -Wang's resin) by a fragment C (9-16) fragment D (17-31) coupling strategy, preferably coupling

Fmoc-Asp(OfBu) ⁹ -Val-Ser(fBu)-Ser(fBu)-Tyr(fBu)-Leu-Glu(OfBu)-Gly ¹⁶ -OH (fragment 2) with

H-Gln(Trt) ¹⁷ -Ala-Ala-Lys(Pal-Glu-OfBu)-Glu-(OfBu)-Phe-Ile-Ala-Trp(NmBoc)-Leu-Val-Arg(Pbf)-Gly-Arg (Pbf)-Gly ³¹ -Wang's resin, fragment 1, a specified fragment D (17-31).

Then, the α-amino protected fragment B (9-31) thus obtained is treated to cleave the Fmoc group under basic conditions as described above to give fragment 3 as defined.

In another preferred embodiment, if preparation of semaglutide is intended, preferred fragment 6 is obtained (ie, H-Asp(OfBu) ⁹ -Val-Ser(tBu)-Ser(tBu)-Tyr(tBu)-Leu- Glu(OtBu)-Gly-Gln(Trt)-Ala-Ala-Lys[(fBuOOC-(CH2)16-CO-yGlu-OfBu)-AEEA-AEEA]-Glu-(OfBu)-Phe-Ile-Ala- Trp(N'"Boc)-Leu-Val-Arg(Pbf)-Gly-Arg(Pbf)-Gly ³¹ -MBH resin) using a fragment C (9-16) fragment D (17-31) docking strategy ), preferably coupling Fmoc-Asp(OfBu) ⁹ -Val-Ser(fBu)-Ser(fBu)-Tyr(fBu)-Leu-Glu(OfBu)-Gly ¹⁶ -OH (fragment 2)

with

H-Gln(Trt) ¹⁷ -Ala-Ala-Lys[(fBuOOC-(CH2)16-CO-yGlu-OfBu)-AEEA-AEEA]-Glu(OfBu)-Phe-Ile-Ala-Trp(NmBoc)- Leu-Val-Arg(Pbf)-Gly-Arg(Pbf)-Gly ³¹ -O-MBH resin, fragment 5, a specified fragment D(17-31).

Then, the α-amino protected fragment B (9-31) thus obtained is treated to cleave the Fmoc group under basic conditions as described above to give fragment 6 as defined above.

As an example for the preparation of fragment B, the preparation of fragment 3 and fragment 6 are described in the examples of the present disclosure, including the preparation of fragment 1, fragment 5 and fragment 2. Analogously, all defined fragment Bs above are prepared by coupling a suitable fragment C and a suitable fragment D as defined above.

The most preferred embodiment of the method for preparing liraglutide is a method according to claim 1 wherein: fragment A is His-Ala-Glu-Gly-Thr-Phe-Thr-Ser(y zzpro), referred to as fragment A(1- 8), more preferably Boc-His(Trt) ¹ -Ala-Glu(OtBu)-Gly-Thr(tBu)-Phe-Thr(tBu)-Ser(^MeMepro) ⁸ -OH, also referred to as fragment 4; and fragment B is Asp-Val-Ser-Ser-Tyr-Leu-Glu-Gly-Gln-Ala-Ala-Lys(Pal-Glu)-Glu-Phe-Ile-Ala-Trp-Leu-Val-Arg- Gly-Arg-Gly-resin, designated fragment B(9-31),

more preferably H-Asp(OfBu) ⁹ -Val-Ser(fBu)-Ser(fBu)-Tyr(fBu)-Leu-Glu(OfBu)-Gly ¹⁶ -Gln(Trt)-Ala-Ala-Lys(Pal- Glu-OfBu)-Glu(OfBu)-Phe-Ile-Ala-Trp(N'"Boc)-Leu-Val-Arg(Pbf)-Gly-Arg(Pbf)-Gly ³¹ -Wang resin, also called fragment 3.

Most preferred in this configuration is that fragment C is Asp-Val-Ser-Ser-Tyr-Leu-Glu-Gly, designated fragment C(9-16), more preferably Fmoc-Asp(OfBu) ⁹ -Val-Ser (fBu)-Ser(fBu)-Tyr(fBu)-Leu-Glu(OtBu)-Gly ¹⁶ -OH, also referred to as fragment 2; and

fragment D is Gln ¹⁷ -Ala-Ala-Lys(Pal-Glu)-Glu-Phe-Ile-Ala-Trp-Leu-Val-Arg-Gly-Arg-Gly ³¹ -resin, also called fragment D (17- 31),

more preferably H-Gln(Trt) ¹⁷ -Ala-Ala-Lys(Pal-Glu-OfBu)-Glu-(OfBu)-Phe-Ile-Ala-Trp(N'nBoc)-Leu-Val-Arg(Pbf) -Gly-Arg(Pbf)-Gly ³¹ -Wang's resin, also called fragment 1.

The most preferred embodiment of the method for preparing semaglutide is a method according to claim 1 wherein,

fragment A is His-Aib-Glu-Gly-Thr-Phe-Thr-Ser(y zzpro), designated fragment A(1-8), more preferably Boc-His(Trt) ¹ -Aib-Glu(OfBu)- Gly-Thr(fBu)-Phe-Thr(fBu)-Ser(^MeMepro) ⁸ -OH, also called fragment 7; and fragment B is Asp-Val-ser-ser-Tyr-Leu-Glu-Gly-Gln-Ala-Ala-Lys[(ROOC-(CH2)16-CO-yGlu-OR')-AEEA-AEEA]- Glu-Phe-Ile-Ala-Trp-Leu-Val-Arg-Gly-Arg-Gly-resin, designated fragment B (9-31),

more preferably H-Asp(OfBu) ⁹ -Val-Ser(fBu)-Ser(fBu)-Tyr(fBu)-Leu-Glu(OtBu)-Gly-Gln(Trt)-Ala-Ala-Lys[(fBuOOC- (CH2)16-CO-yGlu-OfBu)-AEEA-AEEA]-Glu-(OfBu)-Phe-Ile-Ala-Trp(N'nBoc)-Leu-Val-Arg(Pbf)-Gly-Arg(Pbf )-Gly ³¹ -resin from MBH, also called fragment 6.

Most preferred in this configuration is that fragment C is Asp-Val-Ser-Ser-Tyr-Leu-Glu-Gly, designated fragment C(9-16), more preferably Fmoc-Asp(OfBu) ⁹ -Val-Ser (fBu)-Ser(fBu)-Tyr(fBu)-Leu-Glu(OtBu)-Gly ¹⁶ -OH, also referred to as fragment 2; and

fragment D is Gln ¹⁷ -Ala-Ala-Lys[(ROOC-(CH2)16-CO-yGlu-OR')-AEEA-AEEA]-Glu-Phe-Ile-Ala-Trp-Leu-Val-Arg- Gly-Arg-Gly ³¹ -resin, also referred to as fragment D(17-31), wherein R and R', the same or different, are carboxylic acid protecting groups, preferably esters,

more preferably H-Gln(Trt) ¹⁷ -Ala-Ala-Lys[(tBuOOC-(CH2)16-CO-yGlu-OfBu)-AEEA-AEEA]-Glu(OfBu)-Phe-Ile-Ala-Trp(N ,nBoc)-Leu-Val-Arg(Pbf)-Gly-Arg(Pbf)-Gly ³¹ -O-resin from MBH, fragment 5.

The convergent fragment-based method for the preparation of GLP-1 analogues described above provides the final desired resin-bound peptide, which is treated for a simultaneous or sequential deprotection and/or cleavage step to obtain the crude peptide, which optionally further purified.

Abbreviations

SPPS Solid Phase Peptide Synthesis

LPPS Liquid Phase Peptide Synthesis

CSPPS Convergent Solid Phase Peptide Synthesis

CTC 2-chloro-trityl chloride

MBH 4-Methylbenzhydryl

4-MeO-BH 4-Methoxybenzhydryl

AEEA 2-[2-(2-aminoethoxy)ethoxy]acetyl

Fmoc 9-Fluorenylmethoxycarbonyl

Boc Tere-butyloxycarbonyl

Trt Trityl(triphenylmethyl)

tBu Tere-butyl

Pbf 2,2,4,6,7-Pentamethyl-dihydrobenzofuran-5-sulfonyl

eq. Equivalent

h Hour/s

min Minute/s

HPLC High Performance Liquid Chromatography

DIEA/DIPEA N,N-Diisopropylethylamine

DBU 1,8-Diazabicyclo[5.4.0]undec-7-ene

DMAP 4-Dimethylaminopyridine

TFA Trifluoroacetic acid

TIS Triisopropylsilane

Ac ² O Acetic anhydride

DMF N,N-Dimethylformamide

DMA N,N-Dimethylacetamide

DCM Dichloromethane

THF Tetrahydrofuran

NMP N-Methyl-2-pyrrolidinone

MTBE Methyl-ferc-butyl ether

DIPE Diisopropyl ether

MeOH Methanol

HFIP Hexafluoro-2-propanol

TBDMS Tert-butyl-dimethyl-silyl

DIC N,N'-Diisopropylcarbodiimide

DCC N,N'-Dicyclohexylcarbodiimide

EDC N-(3-dimethylaminopropyl)-N'-ethylcarbodiimide

HOBt 1-Hydroxybenzotriazole

HOAt 1-Hydroxy-7-azabenzotriazole

NHS N-Hydroxysuccinimide

TBTU O-(benzotriazol-1-yl)-N,N,N',N'-tetramethyluronium tetrafluoroborate

O-(benzotriazol-1-yl)-N,N,N',N'-tetramethyluronium hexafluorophosphate HBTU

HATU 2-(7-aza-1H-benzotriazol-1 -yl)-1,1,3,3-tetramethyluronium hexafluorophosphate

PyBOP (Benzotriazol-1 -yloxy)tripyrrolidinophosphonium hexafluorophosphate

OxymaPure 2-cyano-2-hydroxyimino-ethyl acetate

Oxyma-B 5-(Hydroxyimino)1,3-dimethylpyrimidine-2,4,6-(1H,3H,5H)-trione

TX-100 Triton X-100, also called polyethylene glycol tert-octylphenyl ether

examples

Detailed experimental parameters suitable for the preparation of liraglutide and semaglutide according to the present invention are provided by the following examples, which are intended to be illustrative and not limiting of all possible embodiments of the invention.

Example 1

Liraglutide preparation

Step 1. Preparation of fragment 1, a fragment D (17-31),

H-Gln(Trt)17-Ala-Ala-Lys(Pal-Glu-OfBu)-Glu(OfBu)-Phe-Ile-Ala-Trp(N'nBoc)-Leu-Val-Arg(Pbf)-Gly- Arg(Pbf)-Gly31-Wang's resin

Synthesis of peptide fragment 1 was carried out at room temperature by stepwise Fmoc SPPS using 0.25 g of Wang's resin (0.75 mmol/g). After swollen the resin in 2 ml of DMF, Fmoc-Gly-OH, DIC and DMAP (4, 2 and 0.1 eq. respectively, based on resin loading) in DMF were added. After 1h, unreacted hydroxyl groups on the resin were capped by treatment with DMF/DIPEA/Ac2O (v/v/v 5/2/1) mixture for 30 minutes. Fmoc deprotection was performed with 20% piperidine solution in DMF (2*2 mL, 5 min and 15 min) and the resin was washed with DMF (4*2 mL). The degree of substitution was checked by measuring the UV adsorption of the collected solution after Fmoc deprotection and found to be 0.53 mmol/g. Fmoc-protected amino acids (4-fold excess relative to resin loading) were preactivated with the mixture of DIC (4 eq) and ethyl 2-cyano-2-(hydroxyimino)acetate (OxymaPure, 4 eq) and subsequently they were coupled with the resin in 60 min. Completion of coupling was monitored by ninhydrin test. In case of incomplete reaction, the coupling was repeated. For the introduction of Fmoc-Lys(Pal-Glu-OfBu)-OH into the peptide sequence double coupling was carried out using 3 eq. and 1.5 eq. From palmitoylated dipeptide (9 h and 5 h, respectively). At the end of each coupling, unreacted peptide chains were acylated with acetic anhydride (10 eq) in the presence of DIPEA (10 eq). Intermediate Fmoc deprotections were carried out using 20% piperidine solution in DMF (2*2 mL, 5 min and 15 min) followed by washing the resin with DMF (4*2 mL). After completion of the synthesis, the resin was washed with DCM and dried.

Step 2. Preparation of fragment 2, a fragment C(9-16),

Fmoc-Asp(OfBu) ⁹ -Val-Ser(fBu)-Ser(fBu)-Tyr(fBu)-Leu-Glu(OfBu)-Gly ¹⁶ -OH

Synthesis of the peptide fragment was carried out by stepwise SPPS using 2-chlorotrityl chloride resin (CTC resin) (0.25 g, 1.6 mmol/g). After swelling the resin using 3 mL of DCM, Fmoc-Gly-OH and DIEA (0.8 and 3 eq., based on resin loading) in DCM were added. The reaction mixture was stirred for 1 hour and the solvent was removed by filtration. Unreacted sites on the resin were capped using a mixture of DCM/DIPEA/MeOH (v/v/v 17/2/1) for 30 minutes. After washing with DCM (3*5 ml) the resin was treated with a mixture of DCM/DIPEA/Ac2O (v/v 5/2/1) for 30 minutes. Resin loading was checked by measuring the UV adsorption of the solution after Fmoc deprotection and found to be of 1.08 mmol/g. Fmoc-Glu(OfBu)-OH, Fmoc-Leu-OH, Fmoc-Tyr(fBu)-OH, Fmoc-Ser(fBu)-OH, Fmoc-Ser(fBu)-OH, Fmoc-Val-OH were then preactivated. and Fmoc-Asp(OfBu)-OH (3-fold excess over resin loading) by DIC (3-fold excess) and OxymaPure (3-fold excess) and sequentially coupled to the resin in 60 min. Intermediate Fmoc deprotections were carried out using 20% piperidine solution in DMF (2 x 2 ml, 5 min and 15 min) followed by washing the resin with DMF (4 * 2 ml). After completion of the synthesis the resin was washed with DCM. The protected peptide was cleaved from the resin by treatment with 1 ml of 1% TFA in DCM for 2 min and the solution was filtered through 2 ml of 10% pyridine solution in methanol. The procedure was repeated 4 times. The resin was then washed with DCM (3*2ml), methanol (3*2ml) and the combined solutions concentrated to 5% volume. The protected octapeptide was precipitated by adding water under an ice bath, filtered, washed several times with water and dried. Yield: 330 mg (90%), HPLC purity 91%

Stage 3-a. Condensation of fragments 1 and 2: preparation of fragment 3, a fragment B (9-31),

H-Asp(OfBu)9-Val-Ser(fBu)-Ser(fBu)-Tyr(fBu)-Leu-Glu(OtBu)-Gly-Gln(Trt)-Ala-Ala-Lys(Pal-Glu-OfBu )-Glu(OfBu)-Phe-Ile-Ala-T rp(N'"Boc)-Leu-Val-Arg(Pbf)-Gly-Arg(Pbf)-Gly31-Wang's resin

Peptide fragment 1 obtained in step 1 was swollen in 2 ml of DMF. Protected fragment 2 (180 mg, 0.13 mmol, 1 eq.) was dissolved in 7 mL DMF, preactivated with DIC (16 mg, 0.13 mmol), OxymaPure (40 mg, 0.27 mmol) and was coupled with fragment 1. The reaction was stirred at 40°C and monitored with HPLC. Another 1 eq. of preactivated peptide fragment 2 (180 mg, 0.13 mmol) at 40°C and stirred to complete condensation. The resin was then washed with DMF (2 * 2 ml) and residual unreacted amino function was capped with 10% Ac ² O solution in DMF (2 ml, 30 min).

Finally, the Fmoc protecting group was removed with 20% piperidine solution in DMF (2 * 2 ml, 5 min and 15 min) with subsequent washing of the resin with DMF (3 * 2 ml).

Stage 3-b. Condensation of fragments 1 and 2 in the presence of a detergent: preparation of fragment 3, a fragment B (9-31),

H-Asp(OfBu)9-Val-Ser(fBu)-Ser(fBu)-Tyr(fBu)-Leu-Glu(OfBu)-Gly-Gln(Trt)-Ala-Ala-Lys(Pal-Glu-OfBu )-Glu(OfBu)-Phe-Ile-Ala-T rp(N'"Boc)-Leu-Val-Arg(Pbf)-Gly-Arg(Pbf)-Gly31-Wang's resin

Peptide fragment 1 obtained in step 1 was swelled in 2 ml of a mixture of DMF:DCM 50:50 v/v 1% Triton X-100 (TX-100). Protected fragment 2 (267 mg, 0.195 mmol, 1.5 eq.) was dissolved in 5.3 mL of 1% DMF:DCM 50:50 v/v TX-100, preactivated with DIC (24.6 mg , 0.195 mmol), OxymaPure (27.7 mg, 0.195 mmol) and coupled with fragment 1. The reaction was shaken at 40°C and monitored by HPLC. The resin was then washed with 2 ml of DMF:DCM 50:50 v/v 1% TX-100 and with DMF (2*2 ml). Residual unreacted amino function was capped with 10% Ac ² O solution in DMF (2 mL, 30 min).

Finally, the Fmoc protecting group was removed with 20% piperidine solution in DMF (2 * 2 ml, 5 min and 15 min) with subsequent washing of the resin with DMF (3 * 2 ml).

Step 4. Preparation of fragment 4, a fragment A (1-8),

Boc-His(Trt)1-Ala-Glu(OfBu)-Gly-Thr(fBu)-Phe-Thr(fBu)-Ser(vMeMepro)8-OH

Peptide fragment 4 was prepared in a similar manner as described in step 2. First, binding of Fmoc-Thr(fBu)-Ser(^MeMepro) ⁸ -OH (0.8 eq.) to the resin of CTC (0.25 g, 1.6 mmol/g) in the presence of DIEA (3 eq) in DCM was carried out to give the loading of 0.96 mmol/g. Fmoc-Phe-OH, Fmoc-Thr(fBu)-OH, Fmoc-Gly-OH, Fmoc-Glu(OfBu)-OH, Fmoc-Ala-OH, and Boc-His(Trt)-OH were then consecutively coupled ( 3-fold excess with respect to resin loading) to the resin within ⁶⁰ min after preactivation with DIC (3 eq.) and OxymaPure (3 eq.). The first Fmoc deprotection was carried out using a 20% solution of piperidine in ^DmF (3 x 2 ml, 3 min) with subsequent washing of the resin with DMF (4 x ² ml). For the deprotection of Fmoc from the other amino acids, two treatments with 20% piperidine solution in DMF (5 min and 15 min) were performed. The protected octapeptide was cleaved from the resin using the same procedure as previously described (see step 2). Yield: 270 mg (80%), HPLC purity 87%

Stage 5-a. Condensation of fragments 3 and 4: liraglutide preparation (1-31)

The Fmoc-deprotected peptidyl resin-supported fragment 3 obtained in step 3 (3-a or 3-b) was swelled in 2 ml of DMF. Protected fragment 4 (272 mg, 0.20 mmol, 1.5 eq.) was dissolved in 7 mL DMF, preactivated with DIC (16 mg, 0.13 mmol), OxymaPure (40 mg, 0.27 mmol ) and ligated with fragment 3. The reaction was stirred at 40°C and monitored with HPLC. Another 1.5 eq. of preactivated fragment 3 (270 mg, 0.20 mmol) and stirred at 40°C to complete the condensation. The resin was then washed with DMF (2 x ² mL) and the residual unreacted amino function was capped with 10% Ac2O solution in DMF (2 mL, 30 min). Then the resin was finally washed twice with DCM (2 ml) and dried.

Stage 5-b. Condensation of fragments 3 and 4 in the presence of a detergent: liraglutide preparation (1-31) The Fmoc-deprotected resin-supported fragment 3 obtained in step 3 (3-a or 3-b) was swelled in 2 ml of a DMF:Dc M mixture 50:50 v/v 1% TX-100. Protected fragment 4 (270 mg, 0.195 mmol, 1.5 eq.) was dissolved in 2.7 mL of 1% DMF:DCM 50:50 v/v TX-100, preactivated with DIC (25 mg, 0.195 mmol), OxymaPure (28 mg, 0.195 mmol) and coupled with fragment 3. The reaction was stirred at 40°C and monitored with HPLC. The resin was then washed with 2 ml of DMF:DCM 50:50 v/v 1% TX-100 and with DMF (2 x ² ml). Residual unreacted amino function was capped with 10% Ac2O solution in DMF (2 mL, 30 min). Then the resin was finally washed twice with DCM (2 ml) and dried.

Stage ⁶ . Cleavage and purification of liraglutide

The dried peptidyl resin obtained in step 5-a was suspended in 2 ml of the TFA/water/phenol/TIS mixture (v/v/v/v 88/5/5/2) and stirred for 4 h. . The resin was then filtered off and washed with 1 ml of TFA. Solutions were collected and the crude peptide was precipitated in 5 ml of MTBE. The precipitate was washed several times with MTBE and dried to obtain crude liraglutide in an overall yield of 375 mg (75%) and an HPLC purity of 81%. Crude liraglutide was purified with 0.1% TFA/water mobile phase A and 0.1% TFA/acetonitrile mobile phase B. Fractions with purity greater than 99.0% were collected and lyophilized to give purified liraglutide. HPLC purity: 99.4%, yield 112 mg (30%).

Example 2

Semaglutide preparation

Step 1. Preparation of fragment 5, a fragment D (17-31)

H-Gln(Trt) ¹⁷ -Ala-Ala-Lys[(fBuOOC-(CH2)16-CO-YGlu-OfBu)-AEEA-AEEA]-Glu(OfBu)-Phe-Ile-Ala-Trp(NinBoc)- Leu-Val-Arg(Pbf)-Gly-Arg(Pbf)-Gly ³¹ -O-MBH resin

Synthesis of peptide fragment 5 was carried out at room temperature using 0.5 g of MBH H-Gly-O-resin (0.60 mmol/g loading). The resin was swollen in 3 ml DMF and used for Fmoc SPPS in stages. Fmoc-protected amino acids (2-fold excess relative to resin loading) were preactivated with the mixture of DIC (2 eq.) and OxymaPure (2 eq.) for 3 min and sequentially coupled to the resin in 90 min. . Arg, Val, Trp, Ala ²⁴ , Phe and Gln ligations were repeated with 1 eq. of the coupling mixture for 60 min. The introduction of the lipid side chain into the peptide sequence was carried out using 1.5 eq. of Fmoc-Lys[(tBuOOC-(CH2)16-CO-yGlu-OtBu)-AEEA-AEEA]-OH, which was preactivated with 1.5 eq. of DIC and 1.5 eq. of OxymaPure and coupled in 6 h at 40°C. Completion of coupling was monitored by ninhydrin test. At the end of each coupling, unreacted peptide chains were acetylated with acetic anhydride (3 eq.) in DMF and washed with DMF (3*5 mL). Intermediate Fmoc deprotections were carried out using 20% piperidine solution in DMF (2*5 mL, 10 min) followed by washing the resin with DMF (4*5 mL). After completion of the synthesis, the resin was washed with DCM and dried.

Step 2. Preparation of fragment 2, a fragment C(9-16),

Fmoc-Asp(OtBu) ⁹ -Val-Ser(fBu)-Ser(fBu)-Tyr(fBu)-Leu-Glu(OtBu)-Gly ¹⁶ -OH See Step 2 of Example 1.

Step 3. Condensation of fragments 5 and 2 in the presence of a detergent: preparation of fragment 6, a fragment B (9-31),

H-Asp(OfBu) ⁹ -Val-Ser(fBu)-Ser(fBu)-Tyr(fBu)-Leu-Glu(OtBu)-Gly-Gln(Trt)-Ala-Ala-Lys[(tBuOOC-(CH2 )16-CO-yGlu-OfBu)-AEEA-AEEA]-Glu-(OfBu)-Phe-Ne-Ala-Trp(N'"Boc)-Leu-Val-Arg(Pbf)-Gly-Arg(Pbf) -Gly ³¹ -MBH resin

Fmoc-deprotected resin with peptide fragment 5 obtained in step 1 was swelled in 4 ml of a mixture of 1% DMF:DCM 50:50 v/v TX-100 at 40°C. The solution of protected peptide fragment 2 (765 mg, 0.56 mmol, 2 eq.) in 7 mL of the 1% DMF:DCM 50:50 v/v TX-100 mixture was preactivated with DIC (16 mg, 0.56 mmol) and OxymaPure (88 mg, 0.56 mmol) at 40°C for 15 min, then added to the resin and the reaction mixture was stirred for 3.5 h at 40°C. Completion of coupling was monitored by ninhydrin test. Unreacted peptide chains were acetylated with 3 eq. of acetic anhydride in DMF and then the peptide resin was washed with DMF (3*5 ml). At the end, the Fmoc protecting group was removed with 20% piperidine solution in DMF (2*5 mL, 10 min) with subsequent washing of the resin with DMF (4*5 mL).

Step 4. Preparation of fragment 7, a fragment A (1-8),

Boc-His(Trt)1-Aib-Glu(OfBu)-Gly-Thr(fBu)-Phe-Thr(tBu)-Ser(^MeMepro)8-OH

The synthesis of peptide fragment 7 was carried out using 1 g of CTC resin (1.6 mmol/g loading). After swollen the resin in 10 mL of DCM, a solution of Fmoc-Thr(fBu)-Ser(^MeMepro)8-OH and DIEA (0.8 and 3 eq., based on resin loading) was added. ) in 5 mL of DCM. The reaction mixture was stirred for 16 hours and the solvent was removed by filtration. Unreacted sites on the resin were capped using a solution of MeOH (3 eq) and DIEA (3 eq) in 5 ml DCM for 15 min. After washing with DCM (2 x 10 ml) the resin was treated with a solution of acetic anhydride (3 eq.) and DIEA (3 eq.) in 10 ml of DCM for 15 minutes, washed with DMF (3 x ¹⁰ ml), DCM (3 x 10 ml) and dried. Resin loading was checked by measuring the UV adsorption of the solution after Fmoc deprotection and found to be 1.3 mmol/g. Fmoc-Phe-OH, Fmoc-Thr(fBu)-OH, Fmoc-Gly-OH, Fmoc-Glu(OfBu)-OH, Fmoc-Ala-OH and Boc-His(Trt)-OH were then preactivated (2 eq .) by 2 eq. of DIC and 2 eq. of OxymaPure for 3 min and sequentially coupled to the resin in 90 min. In the case of His, Oxyma-B replaced OxymaPure and the preactivation time was reduced to 1 min at 0°C. Intermediate Fmoc deprotections were carried out using 20% piperidine solution in DMF (2 x ¹⁰ mL, 10 min) followed by washing the resin with DMF (4 x 10 mL). After completion of the synthesis the resin was washed with DCM (2 x 10 ml) and dried. The protected peptide was cleaved from the resin by treatment with 3 ml of 1.5% TFA in DCM for 2 min and the solution was filtered through 3 ml of 10% pyridine solution in methanol. The procedure was repeated 4 times and the combined solutions were concentrated to 30% volume. The protected peptide was precipitated by adding water under an ice bath, filtered, washed several times with water and dried. Yield: 960 mg (77%).

Step 5. Condensation of fragments 6 and 7: preparation of semaglutide (1-31)

The Fmoc-deprotected peptidyl resin-supported fragment 6 obtained in step 3 was swelled in 2 ml of the mixture of DMF:DCM 50:50 v/v 1% TX-100 at 40°C. Dissolution of protected peptide fragment 7 (780 mg, 0.56 mmol, 2 eq.) in 7 mL of the 1% DMF:DCM 50:50 v/v TX-100 mixture was preactivated with DIC (16 mg, 0.56 mmol) and OxymaPure (88 mg, 0.56 mmol) at 40°C for 15 min. The coupling mixture was then added to the resin and the reaction mixture was stirred for 3.5h at 40°C. Completion of coupling was monitored by ninhydrin test. Unreacted peptide chains were acetylated with 3 eq. of acetic anhydride in DMF and the peptide resin was washed with DMF (3 x 5 ml), DCM (2 x 5 ml) and dried.

Step 6. Cleavage and purification of semaglutide

The dried peptidyl resin obtained in step 5 was suspended in 5 ml of the TFA/water/phenol/TIS mixture (v/v/v/v 88/5/5/2) and stirred for 1 h at 0° C and 3 h at RT. The resin was then filtered off and washed with 1 ml of TFA. Solutions were collected and the crude peptide was precipitated in 25 ml of DIPE. The precipitate was washed several times with DIPE and dried to obtain crude semaglutide in an overall yield of 0.7 g (61%) and an HPLC purity of 74%. Crude semaglutide was purified with 0.1% TFA/water mobile phase A and 0.1% TFA/acetonitrile mobile phase B. Fractions with purity greater than 99.0% were collected and lyophilized to give purified semaglutide. HPLC purity: 99.1%, yield 196 mg (28%).

Claims (1)

Fragment-based solid phase procedure for the preparation of a GLP-1 analogue of formula I HX1EGTFTSDVSSYLEGQAAK(X2)EFIAWLVRGRG (I) in which X1 is Ala and X2 is N-(1-oxohexadecyl)-L-y-glutamyl, either X1 is a -aminoisobutyric acid (Aib) and X2 is N-(17-carboxy-1-oxoheptadecyl)-L-y-glutamyl-2-[2-(2-aminoethoxy)ethoxy]acetyl-2-[2-(2- aminoethoxy)ethoxy]acetyl, wherein the method comprises a final coupling step between an N-terminal fragment A and a C-terminal fragment B and wherein the N-terminal fragment A has a pseudoproline at the C-terminal reactive site. Method according to claim 1, wherein fragment A is selected from the group consisting of His-Xi-Glu-Gly-Thr-Phe-Thr-Ser(^zzpro) His-X1-Glu-Gly-Thr-Phe-Thr-Ser-Asp-Val-Ser(vz,zpro) His-X1-Glu-Gly-Thr-Phe-Thr-Ser-Asp-Val-Ser-Ser(v z,zpro) His-Xi-Glu-Gly-Thr-Phe-Thr(^zzpro), and His-Xi-Glu-Gly-Thr(^zzpro) where X1 is Ala or Aib, and where (^ z,zpro) indicates that Ser, or Thr, is protected with pseudoproline, according to the formula:

in which Y is hydrogen for Ser and Me for Thr, Z is hydrogen or Me, and R1 is the side chain of the amino acid next to the pseudoproline protected amino acid. Method according to claim 2, wherein fragment A is selected from the group consisting of Boc-His-X1-Glu-Gly-Thr-Phe-Thr-Ser(^MeMepro), Boc-His-X1-Glu-Gly-Thr-Phe-Thr-Ser-Asp-Val-Ser(^MeMepro), Boc-His-X1-Glu-Gly-Thr-Phe-Thr-Ser-Asp-Val-Ser-Ser(VMe,Mepro), Boc-His-Xi -Glu-Gly-Thr-Phe-Thr(^ Me'Mepro) and Boc-His-X1-Glu-Gly-Thr(and Me'Mepro) wherein X1 and (^MeMepro) are as defined in claim 2, and wherein fragment A is coupled to a resin-bound protected peptide fragment B, selected from the group consisting of Asp-Val-Ser-Ser-Tyr-Leu-Glu-Gly-Gln-Ala-Ala-Lys(X ² )-Glu-Phe-Ile-Ala--Trp-Leu-Val-Arg-Gly-Arg-Gly -resin, Ser-Tyr-Leu-Glu-Gly-Gln-Ala-Ala-Lys(X ² )-Glu-Phe-Ile-Ala--Trp-Leu-Val-Arg-Gly-Arg-Gly-resin, Tyr-Leu-Glu-Gly-Gln-Ala-Ala-Lys(X ² )-Glu-Phe-Ile-Ala--Trp-Leu-Val-Arg-Gly-Arg-Gly-resin, Ser-Asp-Val-Ser-Ser-Tyr-Leu-Glu-Gly-Gln-Ala-Ala-Lys(X ² )-Glu-Phe-Ile-Ala--Trp-Leu-Val-Arg-Gly-Arg -Gly-resin and Phe-Thr-Ser-Asp-Val-Ser-Ser-Tyr-Leu-Glu-Gly-Gln-Ala-Ala-Lys(X2)-Glu-Phe-Ile-Ala--Trp-Leu-Val-Arg- Gly-Arg-Gly-resin, in which X2 is as defined in claim 1. The method of claim 3, wherein the preparation of fragment B comprises coupling a fragment C with a fragment D, wherein fragment C is selected from the group consisting of Asp-Val-Ser-Ser-Tyr-Leu-Glu-Gly, Ser-Tyr-Leu-Glu-Gly, Tyr-Leu-Glu-Gly, Ser-Asp-Val-Ser-Ser-Tyr-Leu-Glu-Gly, and Phe-Thr-Ser-Asp-Val-Ser-Ser-Tyr-Leu-Glu-Gly, and in which fragment D is Gln17-Ala-Ala-Lys(X2)-Glu-Phe-Ile-Ala-Trp-Leu-Val-Arg-Gly-Arg-Gly31-resin in which X2 is as defined in claim 1. Process according to any one of claims 1 to 4, wherein X1 is Ala and X2 is N-(1-oxohexadecyl)-L-y-glutamyl. Method according to any one of claims 1 to 4, wherein X1 is Aib and X2 is N-(17 carboxy-1-oxoheptadecyl)-Ly-glutamyl-2-[2-(2-aminoethoxy)ethoxy]acetyl-2-[2-(2-aminoethoxy)ethoxy]acetyl. The method of claim 4, wherein fragment C is selected from the group consisting of Fmoc-Asp(OfBu) ⁹ -Val-Ser(fBu)-Ser(fBu)-Tyr(fBu)-Leu-Glu(OtBu) -Gly ¹⁶ -OH, Fmoc-Ser(fBu)-Tyr(fBu)-Leu-Glu(OtBu)-Gly ¹⁶ -OH, Fmoc-Tyr(fBu)-Leu-Glu(OtBu)-Gly ¹⁶ -OH, Fmoc-Ser(fBu)-Asp(OtBu) ⁹ -Val-Ser(fBu)-Ser(fBu)-Tyr(fBu)-Leu-Glu(OtBu)-Gly ¹⁶ -OH and Fmoc-Phe-Thr(fBu)-Ser(fBu)-Asp(OfBu) ⁹ -Val-Ser(fBu)-Ser(fBu)-Tyr(fBu)-Leu-Glu(OtBu)-Gly ¹⁶ -OH, and in which fragment D is H-Gln(Trt) ¹⁷ -Ala-Ala-Lys(Pal-Glu-OfBu)-Glu(OfBu)-Phe-Ile-Ala-Trp(NmBoc)-Leu-Val-Arg(Pbf)-Gly-Arg( Pbf)-Gly ³¹ -Wang resin, or H-Gln(Trt) ¹⁷ -Ala-Ala-Lys[(fBuOOC-(CH2)16-CO-yGlu-OfBu)-AEEA-AEEA]-Glu(OfBu)-Phe-Ne-Ala-Trp(NmBoc)- Leu-Val-Arg(Pbf)-Gly-Arg(Pbf)-Gly ³¹ -O-resin from MBH. Method according to claim 7, wherein fragment C is Fmoc-Asp(OfBu) ⁹ -Val-Ser(fBu)-Ser(fBu)-Tyr(fBu)-Leu-Glu(OtBu)-Gly ¹⁶ -OH and fragment D is H-Gln(Trt) ¹⁷ -Ala-Ala-Lys(Pal-Glu-OfBu)-Glu(OfBu)-Phe-Ile-Ala-Trp(NmBoc)-Leu-Val-Arg(Pbf)-Gly-Arg( Pbf)-Gly ³¹ -Wang resin, or H-Gln(Trt) ¹⁷ -Ala-Ala-Lys[(fBuOOC-(CH2)16-CO-yGlu-OtBu)-AEEA-AEEA]-Glu(OfBu)-Phe-Ne-Ala-Trp(NmBoc)- Leu-Val-Arg(Pbf)-Gly-Arg(Pbf)-Gly ³¹ -O-resin from MBH. Method according to claim 3, wherein fragment A is Boc-His-Ala-Glu-Gly-Thr-Phe-Thr-Ser(V MeMepro) or Boc-His-Aib-Glu-Gly-Thr-Phe-Thr-Ser(V MeMepro) and fragment B is Asp-Val-Ser-Ser-Tyr-Leu-Glu-Gly-Gln-Ala-Ala-Lys(Pal-Glu)-Glu-Phe-Ile-Ala--Trp-Leu-Val-Arg-Gly-Arg- Glyresin, or Asp-Val-Ser-Ser-Tyr-Leu-Glu-Gly-Gln-Ala-Ala-Lys[(ROOC-(CH2)16-CO-yGlu-OR')-AEEA-AEEA]-Glu-Phe-Ile -Ala-Trp-Leu-Val-Arg-Gly-Arg-Gly-resin, respectively wherein R and R', the same or different, are carboxylic acid protecting groups. Method according to claim 9, wherein fragment A is Boc-His(Trt) ¹ -Ala-Glu(OfBu)-Gly-Thr(fBu)-Phe-Thr(fBu)-Ser(v MeMepro) ⁸ -OH or Boc-His(Trt) ¹ -Aib-Glu( OfBu)-Gly-Thr(fBu)-Phe-Thr(fBu)-Ser(and MeMepro) ⁸ -OH, and fragment B is H-Asp(OfBu) ⁹ -Val-Ser(fBu)-Ser(fBu)-Tyr(fBu)-Leu-Glu(OtBu)-Gly-Gln(Trt)-Ala-Ala-Lys(Pal-Glu-OfBu )-Glu(OfBu)-Phe-Ile-Ala-Trp(N /nBoc)-Leu-Val-Arg(Pbf)-Gly-Arg(Pbf)-Gly ³¹ -Wang resin, or H-Asp(OfBu) ⁹ -Val-Ser(fBu)-Ser(fBu)-Tyr(fBu)-Leu-Glu(OtBu)-Gly-Gln(Trt)-Ala-Ala-Lys[(fBuOOC-(CH2 )16-CO-yGlu-OfBu)-AEEA-AEEA]-Glu-(OfBu)-Phe-Ile-Ala-Trp(N mBoc)-Leu-Val-Arg(Pbf)-Gly-Arg(Pbf)-Gly ³¹ -MBH resin, respectively. The method of claim 10, wherein the preparation of fragment B comprises coupling Fmoc-Asp(OfBu) ⁹ -Val-Ser(fBu)-Ser(fBu)-Tyr(fBu)-Leu-Glu(OfBu)- Gly ¹⁶ -OH with H-Gln(Trt) ¹⁷ -Ala-Ala-Lys(Pal-Glu-OfBu)-Glu(OfBu)-Phe-Ile-Ala-Trp(N'nBoc)-Leu-Val-Arg(Pbf)-Gly- Arg(Pbf)-Gly ³¹ -Wang's resin, or H-Gln(Trt) ¹⁷ -Ala-Ala-Lys[(fBuOOC-(CH2)16-CO-yGlu-OfBu)-AEEA-AEEA]-Glu-(OtBu)-Phe-Ne-Ala-Trp(N'"Boc)-Leu-Val-Arg(Pbf)-Gly-Arg(Pbf)-Gly ³¹ -MBH resin, respectively. 12. Process according to any one of the preceding claims, in which the coupling of fragments is carried out in the presence of N,N'-diisopropylcarbodiimide and ethyl 2-cyano-2-hydroxyiminoacetate. 13. Process according to any one of the preceding claims, wherein the coupling of fragments is carried out at a temperature in the range of 10-70°C, more preferably 20-50°C, even more preferably in the range from 35-45°C. 14. Process according to any one of the preceding claims, in which the coupling of fragments is carried out in the presence of a detergent, preferably Triton X-100. The method of any one of the preceding claims, wherein the method further comprises deprotection and cleavage of the peptide from the resin.